The persona you invented to explain a slow quarter
Pipeline was light, so marketing introduced a new ICP. The new ICP didn't exist. The deck had a name for it anyway.
The pipeline chart had been flat for two quarters. In the Monday planning meeting the CMO opened a deck with a new ICP on slide 3: a named buyer persona, a job title, 4 tidy pain points, and a pull-quote that read like ad copy. The persona had everything except a single customer.
The market analysis behind it was a workshop. The marketing team in a room, no actual buyers present, reverse-engineering a customer they wished existed. Over the next six months the team rebuilt messaging, redesigned the site, and ran campaigns at a person the pipeline had never once produced. The slow quarter became three.
The invented persona is the most predictable artifact of marketing under pressure. It shows up when the team can't explain the number and needs a story for the board. The story has one structural defect: it can't be falsified. The persona has no customers yet, and no customers gets read as "we're early" instead of "we're wrong." You can't disprove it for another two quarters, by which point real budget is gone.
Do the empirical version instead. Pull the last 20 closed-won customers. Cluster them by company size, buyer role, trigger event, use case. The cluster that emerges is your ICP. If nothing clusters, you don't have a persona problem, you have a fit problem, and a sharper invented persona won't fix it. It'll hide it.
The bigger trap is that the fiction becomes a forcing function for everyone else. Product builds features the persona "needs." Sales prospects into accounts that match its profile. The hiring plan skills up for its market. None of it is grounded in a real buyer, and all of it spends real money, the same drift that turns a homepage number into a vanity metric. Two quarters in, the company has reorganized around someone who has never signed a contract.
So put a gate in front of it. Before any new persona enters the operating plan, produce 10 to 15 existing closed-won customers who fit it. Not prospects. Not target accounts. Not "logos we're talking to." Paying customers. Under ten, it's fan fiction. Ten or more, it's real, and the repositioning has earned its budget.
A slow quarter usually means something already in the market broke, not that something not yet in the market is missing. Look harder at the customers you have before you invent the ones you wish you had. The whiteboard persona is the better story. The closed-won persona is the cheaper one.